low mem loading of model

app.py

@@ -11,7 +11,7 @@ def load_tokenizer(model_ckpt):
 
 @st.cache(allow_output_mutation=True)
 def load_model(model_ckpt):
-    model = AutoModelForCausalLM.from_pretrained(model_ckpt)
+    model = AutoModelForCausalLM.from_pretrained(model_ckpt, low_cpu_mem_usage=True)
     return model
 
 @st.cache()